cuda.bindings.nvml.device_get_memory_error_counter#

cuda.bindings.nvml.device_get_memory_error_counter( intptr_t device, int error_type, int counter_type, int location_type, ) → unsigned long long#

Retrieves the requested memory error counter for the device.

For Fermi ™️ or newer fully supported devices. Requires NVML_INFOROM_ECC version 2.0 or higher to report aggregate location-based memory error counts. Requires NVML_INFOROM_ECC version 1.0 or higher to report all other memory error counts. Only applicable to devices with ECC. Requires ECC Mode to be enabled. See nvmlMemoryErrorType_t for a description of available memory error types. See nvmlEccCounterType_t for a description of available counter types. See nvmlMemoryLocation_t for a description of available counter locations.

Parameters:

device (intptr_t) – The identifier of the target device.
error_type (MemoryErrorType) – Flag that specifies the type of error.
counter_type (EccCounterType) – Flag that specifies the counter- type of the errors.
location_type (MemoryLocation) – Specifies the location of the counter.

Returns:

Reference in which to return the ECC counter.

Return type:

unsigned long long

Note

On MIG-enabled GPUs, per instance information can be queried using specific MIG device handles. Per instance information is currently only supported for non- DRAM uncorrectable volatile errors. Querying volatile errors using device handles is currently not supported.